Author Information

Kenneth J. Koehler

Abstract

Several methods for analyzing proportions from split-plot and repeated measures experiments are illustrated and compared. One approach simply uses analysis of variance for the usual linear mixed model fit to split-plot and repeated measures experiments. Alternatively, logistic regression analysis is considered and a so-called robust estimate of the covariance matrix is used to adjust for possible correlations among responses. Finally, a quasi-likelihood approach to logistic regression analysis that requires more explicit specification of the covariance structure for the observed proportions is considered. These methods are illustrated with the analyses of data from a repeated measures study of acorn consumption by blue jays and a study of the effects of several environmental factors on nest predation for ground nesting birds.

Keywords

Logistic regression, robust covariance estimation, quasi-likelihood

Creative Commons License


This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

Share

COinS
 
Apr 28th, 9:30 AM

ANALYSIS OF PROPORTIONS FROM SPLIT-PLOT AND REPEATED MEASURES EXPERIMENTS

Several methods for analyzing proportions from split-plot and repeated measures experiments are illustrated and compared. One approach simply uses analysis of variance for the usual linear mixed model fit to split-plot and repeated measures experiments. Alternatively, logistic regression analysis is considered and a so-called robust estimate of the covariance matrix is used to adjust for possible correlations among responses. Finally, a quasi-likelihood approach to logistic regression analysis that requires more explicit specification of the covariance structure for the observed proportions is considered. These methods are illustrated with the analyses of data from a repeated measures study of acorn consumption by blue jays and a study of the effects of several environmental factors on nest predation for ground nesting birds.